Proposed is a one year continuation of "Skills Training and Social Support for Drug Abusers" (NIDA GRANT #5R18DA03013-03). The continuation is requested in order to fully utilize the longitudinal data base constructed in three and one-half years of data collection from subjects first interviewed while in residential drug abuse treatment progrmas (T.C.'s) The data were collected in an experimental test of an aftercare intervention. The data base includes historical, baseline, and 6 and 12 month followup data on 130 subjects who participated in the aftercare experiment. In addition, data are available from the Problem Situation Inventory (PSI), an audiotape recorded roleplay test that assesses subject skills in responding to situations that present a high risk for relapse. The Problem Situation Inventory was administered before and after the experimental intervention. A shortened version was administered at 6 and 12 month followup to assess skill maintenance. Extended analysis of these data sets will advance knowledge in the treatment research field in two areas. First we will construct several different types of drug outcome measures from the data. We will examine the relationships among outcomes from the same subjects using these different measures. As a result, we will develop prescriptive protocols for measuring drug outcomes after treatment. This analysis will facilitate comparison across treatment evaluation studies and will provide an empirical basis for defining appropriate goals for primary drug abuse treatment and aftercare. Second, the PSI data will be analyzed to identify component skills that are predictive of recovery or relapse. The results of the analysis will be used to finalize the component scoring system of the PSI. The final product will be a validated roleplay test which can be used as a diagnostic, clinical intervention and evaluation tool in drug treatment.